检索结果-内蒙古大学图书馆

41st International Conference on Machine Learning, ICML 2024

作者： Wan, Guancheng Tian, Yijun Huang, Wenke Chawla, Nitesh V. Ye, Mang National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China Department of Computer Science University of Notre Dame United States Taikang Center for Life and Medical Sciences Wuhan University Wuhan China

Graph Contrastive Learning (GCL) has emerged as a highly effective self-supervised approach in graph representation learning. However, prevailing GCL methods confront two primary challenges: 1) They predominantly operate under homophily assumptions, focusing on low-frequency signals in node features while neglecting heterophilic edges that connect nodes with dissimilar features. 2) Their reliance on neighborhood aggregation for inference leads to scalability challenges and hinders deployment in real-time applications. In this paper, we introduce S3GCL, an innovative framework designed to tackle these challenges. Inspired by spectral GNNs, we initially demonstrate the correlation between frequency and homophily levels. Then, we propose a novel cosine-parameterized Chebyshev polynomial as low/high-pass filters to generate biased graph views. To resolve the inference dilemma, we incorporate an MLP encoder and enhance its awareness of graph context by introducing structurally and semantically neighboring nodes as positive pairs in the spatial domain. Finally, we formulate a cross-pass GCL objective between full-pass MLP and biased-pass GNN filtered features, eliminating the need for augmentation. Extensive experiments on real-world tasks validate S3GCL proficiency in generalization to diverse homophily levels and its superior inference efficiency. Copyright 2024 by the author(s)

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

The Good, The Bad, and Why: Unveiling Emotions in Generative AI 41

The Good, The Bad, and Why: Unveiling Emotions in Generative...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Li, Cheng Wang, Jindong Zhang, Yixuan Zhu, Kaijie Wang, Xinyi Hou, Wenxin Lian, Jianxun Luo, Fang Yang, Qiang Xie, Xing Microsoft Research Beijing China CAS Institute of Software Beijing China Department of Computer Science William & Mary WilliamsburgVA United States School of Psychology Beijing Normal University Beijing China Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong

Emotion significantly affects our daily behaviors and interactions. Although recent generative AI models, such as large language models, have shown impressive performance in various tasks, it remains unclear whether they truly comprehend emotions and why. This paper aims to address this gap by incorporating psychological theories to gain a holistic understanding of emotions in generative AI models. Specifically, we propose three approaches: 1) EmotionPrompt to enhance the performance of the AI model, 2) EmotionAttack to impair the performance of the AI model, and 3) EmotionDecode to explain the effects of emotional stimuli, both benign and malignant. Through extensive experiments involving language and multi-modal models on semantic understanding, logical reasoning, and generation tasks, we demonstrate that both textual and visual EmotionPrompt can boost the performance of AI models while EmotionAttack can hinder it. More importantly, EmotionDecode reveals that AI models can comprehend emotional stimuli similar to the dopamine mechanism in the human brain. Our work heralds a novel avenue for exploring psychology to enhance our understanding of generative AI models, thus boosting the research and development of human-AI collaboration and mitigating potential risks. Copyright 2024 by the author(s)

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Annolog: A Query Processing Framework for Modelling and Reasoning with Annotated Data

引用

Journal of computers (Taiwan) 2023年第2期34卷 81-97页

作者： Zou, Haochen Wang, Dejian Xiao, Yang Department of Computer Science and Software Engineering Concordia University Montreal Canada

Data annotation is the categorization and labelling of data for applications, such as machine learning, artificial intelligence, and data integration. The categorization and labelling are done to achieve a specific use case in relation to solving problems. Existing data annotation systems and modules face imperfections such as knowledge and annotation not being formally integrated, narrow application range, and difficulty to apply on existing database management applications. To analyze and process annotated data, obtain the relationship between different annotations, and capture metainformation in data provenance and probabilistic databases, in this paper, we design a back-end query processing framework as a supplementary interface for the database management system to extend operation to datasets and boost efficiency. The framework utilizes Java language and the MVC model for development to achieve lightweight, cross-platform, and high adaptability identities. The contribution of this paper is mainly reflected in two aspects. The first contribution is to implement query processing, provenance semiring, and semiring homomorphism over annotated data. The second contribution is to combine query processing and provenance with SQL statements in order to enable the database manager to invoke operations to annotation. © 2023 computer Society of the Republic of China. All rights reserved.

关键词： Query processing

来源：评论

学校读者我要写书评

暂无评论

AI Ethics in Healthcare-A Survey 23

AI Ethics in Healthcare-A Survey

引用

23rd IEEE International Conference on software Quality, Reliability, and Security Companion, QRS-C 2023

作者： Singh, Bhavraaj Rahim, Muhammad Abdul Basit Ur Hussain, Shahid Rizwan, Muhammad Abid Zhao, Jie California State University Department of Computer Science Long BeachCA United States Penn State University Department of Computer Science Behrend PA United States Florida Polytechnic University Department of Computer Science LakelandFL United States Penn State University Department of Computer Science and Software Engineering Behrend PA United States

ISBN: (纸本)9798350359398

In the realm of healthcare, the exponential growth of Artificial Intelligence has precipitated a need to scrutinize its ethical implications. This research undertakes a comprehensive survey to unravel the intricate tapestry of AI ethics within the healthcare landscape. Objective is to delineate the multifaceted challenges that arise from the symbiotic relationship between AI and healthcare and proposing viable solutions for mitigation. A pivotal focus of this study is to bridge the divide between medical practitioners and AI developers, thus addressing a conspicuous research gap. This gap pertains to fostering seamless collaboration between these stakeholders, ensuring that AI systems align with the actual requirements of healthcare providers. The paper explores strategies to establish an effective dialogue, facilitating the design and implementation of ethically sound AI applications. The paper also delves into the moral conundrums engendered by AI's lack of emotional intelligence in sensitive healthcare contexts. The absence of human emotional comprehension has, in certain instances, led to grievous outcomes, necessitating a nuanced approach to machine autonomy. This study advocates for an equilibrium where intelligent machines operate under prudent human oversight, striking a harmonious balance between precision and compassion. Furthermore, the research evaluates prevailing systems and their attendant challenges, emphasizing the advantages of integrating ethically guided, intelligent systems. The paper contemplates governance structures, protocols and strategies to counteract biases inherent in AI algorithms. By dissecting the principles of fairness, accountability and transparency, this study paves the way for a cogent framework that governs AI deployment within healthcare. In essence, this paper charts an uncharted course through the unexplored terrain of AI ethics in healthcare. It not only recognizes the inherent challenges but also underscores the imperative

关键词： Data privacy

来源：评论

学校读者我要写书评

暂无评论

There Is More to Dwell Than Meets the Eye: Toward Better Gaze-Based Text Entry Systems With Multi-Threshold Dwell 25

There Is More to Dwell Than Meets the Eye: Toward Better Gaz...

引用

2025 CHI Conference on Human Factors in Computing Systems, CHI 2025

作者： Mutasim, Aunnoy K Bashar, Mohammad Raihanul Lutteroth, Christof Batmaz, Anil Ufuk Stuerzlinger, Wolfgang Simon Fraser University VancouverBC Canada Department of Computer Science & Software Engineering Concordia University MontrealQC Canada Computer Science University of Bath Bath United Kingdom

ISBN: (纸本)9798400713941

Dwell-based text entry seems to peak at 20 words per minute (WPM). Yet, little is known about the factors contributing to this limit, except that it requires extensive training. Thus, we conducted a longitudinal study, broke the overall dwell-based selection time into six different components, and identified several design challenges and opportunities. Subsequently, we designed two novel dwell keyboards that use multiple yet much shorter dwell thresholds: Dual-Threshold Dwell (DTD) and Multi-Threshold Dwell (MTD). The performance analysis showed that MTD (18.3 WPM) outperformed both DTD (15.3 WPM) and the conventional Constant-Threshold Dwell (12.9 WPM). Notably, absolute novices achieved these speeds within just 30 phrases. Moreover, MTD's performance is also the fastest-ever reported average text entry speed for gaze-based keyboards. Finally, we discuss how our chosen parameters can be further optimized to pave the way toward more efficient dwell-based text entry. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Text Entry

来源：评论

学校读者我要写书评

暂无评论

Enhanced Mask R-CNN blur instance segmentation based on generated adversarial network 2

Enhanced Mask R-CNN blur instance segmentation based on gene...

引用

2nd International Conference on Electronic Information Technology, EIT 2023

作者： Han, Yuehu Chen, Shijie Zhang, Xiaoyan Department of Software Engineering Chongqing University of Posts and Telecommunications Chongqing China Department of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing China

ISBN: (数字)9781510666627

ISBN: (纸本)9781510666610

Aiming at the detection of blur SEM nanoparticle images, a blur instance segmentation network (BL-Mask R-CNN) based on generative adversarial network deblurring convolution block (Deblur) and Mask R-CNN instance segmentation algorithm is proposed. The network uses the enhanced method of the blind deblurring algorithm (DeblurGAN-v2) of the generative confrontation network to preprocess the image, restore the grainy texture details in the image and generate a clear image that is conducive to instance segmentation and detection. The experimental results show that this method effectively improves the detection accuracy of blurred SEM nanoparticle images. Tested on the blur images of the NFFA-EUROPE dataset, the improved BL-Mask R-CNN has compared the detection accuracy of the original Mask R-CNN. The accuracy rate AP has increased from 0.8339 to 0.9613, and achieved good results in the test. © 2023 SPIE.

关键词： Nanoparticles

来源：评论

学校读者我要写书评

暂无评论

Gamification of E-Learning Apps via Acceptance Requirements Analysis 19

Gamification of E-Learning Apps via Acceptance Requirements ...

引用

19th International Conference on Evaluation of Novel Approaches to software engineering, ENASE 2024

作者： Calabrese, Federico Piras, Luca Al-Obeidallah, Mohammed Ghazi Egbikuadje, Benedicta Oghenevoke Alkubaisy, Duaa Department of Information Engineering and Computer Science University of Trento Trento Italy Department of Computer Science Middlesex University London United Kingdom Department of Software Engineering Al Ain University Abu Dhabi United Arab Emirates College of Applied Studies and Community Service Imam Abdulrahman Bin Faisal University Dammam Saudi Arabia

ISBN: (纸本)9789897586965

In the last few years, and particularly during and after the COVID-19 pandemic, E-Learning has become a very important and strategic asset for our society, relevant both for academic and industry settings, involving participants ranging from students to professionals. Different applications have been developed to support E-Learning to be an effective tool, particularly in relation to the software engineering and programming areas. However, in order to be effective, in particular within academic settings, such tools require students to be continuously engaged and motivated to learn both practical and theoretical aspects. The integration of gamification in educational environments has gained considerable prominence as a potential mean to augment students’ motivation and involvement, providing them with immediate feedback and reinforcement, bolstering their sense of accomplishment and motivation to persist in their studies. However, to design gamified applications that can effectively engage and motivate users, as the literature has demonstrated, it is required to consider psychological, sociological, and human behavioural aspects, often referred to as Acceptance Requirements. This study presents a case study, where a Goal Modeling-based, Systematic, Acceptance Requirements Analysis and Gamification Design process, has been applied, by using the Agon framework, to prototype a gamified tool, aiming at engaging students towards learning both theory and practice related to a "Web-Based Mobile App Development" university module. To evaluate our proposed prototype, students were involved to use our proposed gamified prototype. The results show that our gamification solution can engage and motivate students towards learning both theoretical and practical aspects of the module. © 2024 by SCITEPRESS – science and Technology Publications, Lda.

关键词： Requirements engineering

来源：评论

学校读者我要写书评

暂无评论

Evaluating LoRaWAN performance in intentional and unintentional DoS attacks by legacy 900MHz network devices 6

Evaluating LoRaWAN performance in intentional and unintentio...

引用

6th International Conference on Advanced Communication Technologies and Networking, CommNet 2023

作者： Bang, Seokhyeon Jang, Junyoung Ro, Minju Choi, Yuseon Kwon, Doyong Lee, Kangyeon Smith, Anthony Chung-Ang University School of Computer Science and Engineering Seoul Korea Republic of Chonnam National University Department of Software Engineering Gwangju Korea Republic of Kyungpook National University School of Computer Science and Engineering Daegu Korea Republic of Purdue University Department of Computer and Information Technology IN United States

ISBN: (纸本)9798350329391

The deployment of LoRaWAN on the Internet of Things (IoT) has increased since its advent and LoRaWAN now predominates the IoT market over other Low Powered Wide Area Networks (LPWAN). However, since LoRaWAN uses Chirp Spread Spectrum (CSS), it is susceptible to wideband jamming attacks. In this paper, we demonstrate with experiments and concrete numerical results that jamming US915 LoRaWAN frequency is possible by the usual data transmission and reception process of 900MHz Canopy, one of the legacy 900MHz network device. Intentional attack is possible in the same manner. The experiments emulate the real-world environment operated in medical and agriculture industries, in outdoor and indoor conditions, respectively. In addition, this paper introduces and utilizes the novel metric, Jamming Effect (JE), that indicates the network performance of wireless networks that spread the data on air. © 2023 IEEE.

关键词： Risk assessment

来源：评论

学校读者我要写书评

暂无评论

Advancing Efficient Large-Scale Constraint Optimization Problem Solving by Integrating Machine Learning and Automated Reasoning 2

Advancing Efficient Large-Scale Constraint Optimization Prob...

引用

2nd International Workshop on Spatio-Temporal Reasoning and Learning, STRL 2023

作者： Kong, Shufeng School of Software Engineering Sun Yat-sen University Guangzhou China Department of Computer Science Cornell University Ithaca United States

来源：评论

学校读者我要写书评

暂无评论

A Brief Overview of Deep Learning Approaches for IoT Security 4

A Brief Overview of Deep Learning Approaches for IoT Securit...

引用

4th International Conference on Computing, Mathematics and engineering Technologies, iCoMET 2023

作者： Hassaan Khalid, Muhammad Sharif, Hanan Rehman, Faisal Naeem Ullah, Muhammad Shaukat, Shahbaz Maqsood, Hadia Nouman Ali, Chaudhry Hussain, Ayaz Iftikhar, Irfana Lahore Leads University Department of Computer Science & Information Technology Lahore Pakistan Mehran University of Engineering and Technology Szab Department of Software Engineering Khairpur Mir's Pakistan

ISBN: (纸本)9798350335316

The continuous increasing usage of internet devices in many areas of human life is continuously growing and demanding a proper method to protect these IoT devices from cyber-attacks and vulnerabilities. In this aspect, deep learning neural networks are a very helpful and beneficial method to provide protection to internet devices in many aspects. This study's objective is to systematically research and examine the background of research about neural networks tactics functional to different internet-based devices' safety use-cases. We put the influences we looked at into many different groups to show where this research is lacking. This particular research or findings mainly captures the articles related to 'Deep neural networks', 'Cyber risks' and 'IOT' in four databanks, namely science Direct, ACM Digital Library, IEEEXplore and Springer Link. These studies concern 3 main studies, which involve safety aspects, their own Deep Learning network designs, and the occupied datasets. The final round is to find out the loopholes to be examined in the future and vulnerabilities and drawbacks in the IoT security scenario. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：